Yay for R!!

DISCLAIMER - I am not an R-expert…it frustrates the crap out of me. I have to google pretty much everything.

Things to go over today:

1. Getting around in R studio, general R syntax
2. Project files & working directories
3. Inputting data, data types, how to check the structure/type
4. Organizing your data for R
5. Simple plotting bc it's fun
6. Packages galore

Since real data is more fun than fake data, the practice data set for today is about my family’s dogs - here are 2 of these doggos enjoying life. A must for learning R is taking breaks to enjoy life like Tally & Willow! They are pros at taking breaks.

Alt text here

1. Getting around in R studio, general R syntax

Rstudio is like a pretty shell for “base R” - the 4 windows are your script/data, the console, the environment, and the file/plot viewer. You can rearrange them using the button that looks like a window at the top of the screen.

“#” or hashtags tell R “this isn’t code” -it is useful for comments or for turning off certain sections of it

#arrows <- are used to assign something...= can be used too, there is a difference and idk what it is
A <- 6
#you can run code by highlighting it and hitting "run" or by hitting cntrl enter on that line
#you can also type directly into the console, or arrow up in the console to rerun the same line 

B <- 3

#you can use R like a glorified horrible calculator...but knowing this type of syntax is important for other stuff
6+3
## [1] 9
A + B
## [1] 9
C <- A + B
#Another important thing to note in R is that you are typically running a "function" on an "object" 

#functions are generally verbs, and followed by parentheses...like this:

print(A)
## [1] 6
#objects are nouns - things like your data, a model you wrote, etc. A, B, & C are just values, but in a bit we will input a data frame.

2. Project files & working directories

#This function prints your working directory:
getwd()
## [1] "C:/Users/lindsay.malone/Documents/IntroR"
#you can specify file paths, but it's much easier to set your working directory - you can go up to "Session>Set Working Directory>source file location or wherever else you have data saved, or specify:

setwd("~/IntroR")

Using project files is really helpful for keeping things organized! Definitely not crucial though.

Here is a great blog post explaining why project files are useful: https://spatiallychallenged.com/2018/12/07/project-files-and-data-management-for-qgis-and-r/

3. Inputting data, data types, how to check the structure/type

There are LOTS of ways to input data! You can “import dataset” button in the environment tab, or point to the file directly

#pointing to the file

#this line will print out the data, but doesn't save it as anything
read.csv("doggies.csv", sep = ',')
##      name color weight age
## 1   Rosie black     81   2
## 2    Blue black     95   2
## 3 Chelsea brown     90  10
## 4   Tally brown     58   5
## 5  Willow black     40   2
## 6    Jack black     65   4
#by assigning it a name, R remembers the dataset now...note that it doesn't matter what you name things - even though the file is named "doggies" I can reassign it to "puppers" and R doesn't care. 
puppers <- read.csv("doggies.csv")

#this allows you to look at the different data types
str(puppers)
## 'data.frame':    6 obs. of  4 variables:
##  $ name  : chr  "Rosie" "Blue" "Chelsea" "Tally" ...
##  $ color : chr  "black" "black" "brown" "brown" ...
##  $ weight: int  81 95 90 58 40 65
##  $ age   : int  2 2 10 5 2 4

4. Organizing your data for R

R can only handle 1 sheet at a time…(I think). I tend to save excel files with multiple sheets, and save the sheet I want for R as a csv…this is not a great practice, but data management is hard :(

Some big “rules” are: -no spaces -no periods (this is a soft rule) -no weird symbols -if a datapoint is missing, leave the cell blank. SAS wants periods for missing points so this can be a common issue.

there are more…but I cannot think of them right now.

5. simple plotting bc it’s fun You can make some interesting plots without any packages, but it’s tricky. This is nice for getting a quick look at data.

plot(puppers$age)

plot(weight~age, data = puppers)

boxplot(age~color, data=puppers)

6. Packages galore

There is only so much you can do in base R…but there are ENDLESS and constantly growing possibilities with R packages…you have to install them in order to use them the first time, and load them each time you start a new R session.

ggplot is a great one used for making plots…here is a good tutorial for it (I still have it bookmarked!) http://r-statistics.co/Complete-Ggplot2-Tutorial-Part1-With-R-Code.html

library(ggplot2)

ggplot(puppers, aes(x=weight, y=age, color=color)) +
  geom_point(aes(col=color, size=5)) + 
  theme_bw()+
  scale_color_manual(values = c("black"="black",
                                "brown"="sienna4"))

This is just the beginning! Some of my recommendations for continuing to learn R are:

Learning “the tidyverse” - this is a helpful collection of packages that help with data management - ggplot is part of it. - Here is a link to cheat sheets for it: https://www.rstudio.com/resources/cheatsheets/ - There is a LinkedIn Learning course called “Learning the R Tidyverse” with Charlie Joey Hadley that is a great way to get started

Learn to use R markdown - this is something I am currently working on…or trying to, at least.This document is actually an R markdown file!

As anyone who is an expert in this…google is your friend, and asking your actual friends for code is strongly encouraged :) If you need to do something specific and can’t figure it out, just ask!